Journal of Transportation Technologies, 2012, 2, 305-314
http://dx.doi.org/10.4236/jtts.2012.24033 Published Online October 2012 (http://www.SciRP.org/journal/jtts)
Video Based Vehicle Detection and Its Application in
Intelligent Transportation Systems
Naveen Chintalacheruvu, Venkatesan Muthukumar
Department of Electrical and Computer Engineering, University of Nevada, Las Vegas, USA
Received July 30, 2012; revised August 28, 2012; accepted September 7, 2012
ABSTRACT
Video based vehicle detection technology is an integral part of Intelligent Transportation System (ITS), due to its
non-intrusiveness and comprehensive vehicle behavior data collection capabilities. This paper proposes an efficient
video based vehicle detection system based on Harris-Stephen corner detector algorithm. The algorithm was used to
develop a standalonevehicle detection and tracking system that determines vehicle counts and speeds at arterial road-
ways and freeways. The proposed video based vehicle detection system was developed to eliminate the need of com-
plex calibration, robustness to contrasts variations, and better performance with low resolutions videos. The algorithm
performance for accuracy in vehicle counts and speed was evaluated. The performance of the proposed system is equi-
valent or better compared to a commercial vehicle detection system. Using the developed vehicle detection and tracking
system an advance warning intelligent transportation system was designed and implemented to alert commuters in ad-
vance of speed reductions and congestions at work zones and special events. The effectiveness of the advance warning
system was evaluated and the impact discussed.
Keywords: Vehicle Detection; Video and Image Processing; Advance Warning Systems
1. Introduction
The goals of Intelligent Transportation System (ITS) are
to enhance public safety, reduce congestion, improved
travel and transit information, generate cost savings to
motor carriers and emergencies operators, reduce detri-
mental environmental impacts, etc. ITS technologies as-
sist states, cities, and towns nationwide to meet the in-
creasing demands on surface transportation system. The
efficiency of an ITS system is mainly based on the per-
formance and comprehensiveness of the vehicle detec-
tion technology. Vehicle detection and tracking are an
integral part of any vehicle detection technology, since it
gathers all or part of the information that are used in an
effective ITS.
In transportation, vehicle detection system may be de-
fined as a system which is capable of detecting vehicles
and measure traffic parameters such as count, speed, in-
cidents, etc. Also vehicle detection can be used for vari-
ous transportation applications like: autonomous vehicle
guidance, vehicle safety, etc. Vehicle detection by video
cameras is one of the most promising non-intrusive tech-
nologies for large-scale data collection and implementa-
tion of advanced traffic control and management sche-
mes. Vehicle detection is also the basis for vehicle track-
ing. The correct vehicle detection results in better tracking.
Modern computer controlled traffic systems have more
complex vehicle detection requirements than those adop-
ted for normal traffic-actuated controllers for traffic sig-
nals, for which many off-the-shelf vehicle detectors were
designed [1,2]. Many useful and comprehensive parame-
ters like-count, speed, vehicle classification, queue lengths,
volume/lane, lane changes, microscopic and macroscopic
behaviors can be evaluated through video based vehicle
detection and tracking. Autoscope [1] and Iteris [2] are
example of off-the-shelf commercial video based vehicle
detection systems most commonly used in the nation.
This work focuses on developing a real-time vehicle
detection system for low-resolution traffic video feed. The
developed system determines the total and lane based
vehicle counts and average speed of the vehicle for a
given segment of the roadway. There exist many off-the-
shelf commercial video detection systems for vehicles for
various applications like vehicle detection at intersec-
tions, vehicle incident detections, etc. However, vehicle
data collection have been traditionally approached using
one of the following sensor based approaches: radar, li-
dar, loop detectors, microwave sensors, etc. Lately, video
based vehicle data collection systems are being explored
[3] by commercial video detection systems. Moreover,
these systems require extensive calibration and require
user knowledge and expertise to configure these systems.
Also, knowledge of unknown or accurate parameters
C
opyright © 2012 SciRes. JTTs

N. CHINTALACHERUVU, V MUTHUKUMAR
306
(height of the camera) are required to obtain better results.
Also variation in illumination of the video degrades the
efficiency of the system. In order to address the above
drawbacks, we propose a vehicle detection system for ITS
application based on Harris-Stephen Corner Method
(HSCM). The system requires fewer calibrations and is
less immune to illumination changes because of the point
detector and tracking methodologies adopted. The deve-
loped system was evaluated and the performance ana-
lyzed using a set of video feeds (8 sets) of 1 min. interval.
The video feeds vary in illumination (captured during
various times of the day), camera mount height, camera
view angle and region of view. The system was also im-
plemented on an embedded computer platform. The eva-
luations are tabulated and the advantages and accuracy of
our implementation and discussed.
2. Background
Video based object or motion detection and tracking are
two tasks that play a fundamental role in video surveil-
lance systems, transportation systems, military applica-
tions, gaming systems, etc. This section mainly focuses
on the problem of video based vehicle detection and
tracking for ITS applications. Vehicle detection is a pro-
cess of detecting the presence or absence of a vehicle in
the video sequence. Vehicle tracking is defined as find-
ing the location of a vehicle in each frame of the video
sequence. Typically the result of detection is used as ini-
tialization process for tracking. Video based vehicle de-
tection and tracking systems for ITS applications are
performed using: 1) static or moving cameras, 2) single or
multiple cameras, 3) fixed or Pan-Tilt-Zoom (PTZ) ca-
meras. The efficiency of any vehicle detection system is
based on the systems readiness to handle loss of informa-
tion, noise in video, complexity of vehicle motion, vehi-
cle occlusion, shape complexity, illumination changes and
real-time processing.
Vehicle detection and tracking approaches can be bro-
adly classified based on the representation of the ob-
ject/vehicle, detection methods and tracking methods. Re-
presentations of vehicles for detection and tracking in-
clude points, shapes, silhouette, contours, and object mo-
dels [4]. Some of the initial approaches to vehicle detec-
tion and tracking systems involve spatial, temporal or spa-
tio-temporal analysis of video sequences. Vehicle detec-
tion and tracking in general has been performed using
one of the following methodology: point detection and
tracking [5-8], edge detection [9-12], frame differentia-
tion [13-16] thresholding and segmentation followed by
feature extraction [17,18] and matching [19-22] (by cor-
relation or template matching or supervised learning),
and by optical flow methods [23-26]. Point detection and
tracking methods are fast and provide better results for
illumination changes. Edge detection methods employ
morphological edge detection schemes to determine the
object/vehicle. Edge detection techniques are relatively
fast and are less immune to illumination variance. How-
ever, tracking of the vehicle has to be performed by re-
cursive edge detection on subsequent frames. Also, solu-
tions for fixing disconnects in edges and contour irregu-
larities are time consuming and vulnerable to noise. Frame
differencing approaches are relatively fast, but require
either a static background or reference image [15,27] or
frequent updating of the background image [27,28] mak-
ing it not suitable for slow moving vehicles. Feature ex-
traction and matching methods derive dimensions, tex-
tures, color, shapes, etc. of vehicles and matching them
and validate using templates or by correlation. Even
though, these methods are comprehensive and can pro-
vide higher accuracy in differentiating objects, they are
inflexible and are susceptible to intensity variations, sha-
dows and object occlusions. Supervised learning methods
are more suitable for object detection only, time con-
suming (training), and inflexible for location changes.
Optical flow methods encode the temporal displacement
of the pixels during motion and the variation in the spa-
tial structure of the scene. This approach is computa-
tional expensive, vulnerable to noise, perspective distor-
tion, occlusion, static objects and sensitive to the camera
position. Some of the challenges of effective vehicle de-
tection and tracking compared to general object detection
and tracking are: 1) detection and tracking should be fast
due to the relative fast movement of the object/vehicle, 2)
should be independent of the location, camera region of
view, camera resolution and mounted height, illumine-
tion, shadows and occlusion, 3) require less calibrations
to determine traffic flow data, and 4) maximize true de-
tections and minimize false detections. On the other hand,
vehicle detection and tracking provide less challenge com-
pared to general object tracking in: 1) vehicle direction
and displacement is common and proximity-uniform, 2)
vehicles shape and features are distinct from background
and active objects (pedestrians), 3) motion of vehicles are
bound to the roadway area (smaller frame size for pro-
cessing), and 4) object detection by matching and super-
vised learning are overwhelming and is only required for
vehicle classification.
The main goal of this work was to develop real-time
embedded vehicle detection and tracking system from
low resolution CCTV camera feeds to determine vehicle
count and flow parameters that could be used for various
ITS applications. Given the above realistic goals, point
detection and tracking methodology was selected for our
vehicle detection and tracking system for the following
reasons: 1) extremely fast and can detect multiple objects
in a single frame, 2) better performance with illumination

N. CHINTALACHERUVU, V MUTHUKUMAR 307
variance, 3) since vehicle direction and displacement is
common and proximity-uniform, iterative point tracking
across multiple frames are quick, 4) camera region of view
and camera positioning height are not required, and 5)
can handle partial occlusion.
Point detectors are used to find interest points in im-
ages. Interest points have classically been used for mo-
tion detection and vehicle tracking. Also, interest point
detectors have been proved to be invariant to illumina-
tion changes and camera viewpoint. The commonly used
interest point detectors in literature are: Moravec’s op-
erator [5], Harris-Stephen corner interest point detector
[6], KLT detector [7], and SIFT detector [8]. Moravec’s
operator computes the image intensity variation in a 4 × 4
patch only for discrete set of shifts at every 45 degrees.
Moravec’s operator fails to detect an edge/interest point if
the edge is in the direction of its neighbors. Harris- Ste-
phen improved upon Moravec’s operation by considering
the following: 1) all possible small shifts are covered by
performing an analytic expansion about the shift origin, 2)
reduce noise by considering a smooth circular Gaussian
window, 3) compute first order image derivatives in x
and y directions to highlight the directional intensity vari-
ations, then a second order moment matrix, which en-
codes this variation for each pixel in a small neighbor-
hood. The interest points are evaluated by determining
the determinant and trace of this matrix and the interest
point are derived after thresholding the interest point con-
fidence value after applying non-maxima suppression.
Point tracking methods determine the correspondence
of interest points across frames. Point correspondence
methods can be broadly classified as deterministic and
statistical methods [4]. Deterministic methods use quail-
tative motion heuristics to constrain the correspondence
problem. Statistical methods use explicit object features,
parameters, and uncertainties into consideration to deter-
mine correspondence. Deterministic methods define a cor-
respondence cost with a set of motion constraints asso-
ciated with each object between frames. Some of the
motion constraints used in literature for point tracking
include: proximity, maximum velocity, smooth motion,
common motion, rigidity and proximal-uniformity [4].
The vehicle detection and tracking system implemented
in this work is compared with Autoscope [29], a comer-
cial video based vehicle detection system. The Auto-
scope vehicle detection system is based on background
frame differencing [30] and inter-frame differencing for
vehicle detection and edge detection and centroid corre-
spondence method for tracking. The system also has sig-
nificant built-in heuristics for shadows elimination and
for detection under various weather conditions. Also, height
of the camera position significantly influences the detec-
tion and tracking accuracy. The developed vehicle detec-
tion and tracking system is based on interest point detec-
tion and tracking using Harris-Stephen corner detector
and point correspondence to determined the displacement
shift in pixels corresponding to the vehicle travel. The
developed system was used as vehicle detector for the
projects, “Testing and Evaluation of the Effectiveness of
Advanced Technologies for Work Zones” and “Test Queue
Detection Systems for Preventing Accidents in Nevada”,
which were sponsored by Nevada Department of Trans-
portation. The vehicle detector system was developed us-
ing Harris-Stephen corner detection algorithm using
OpenCV library on a Arcom’s Olympus Windows XP Em-
bedded development kit running WinXPE operating sys-
tem. The system developed was used to detect vehicle
flow (count and speed) to determine congestion at work
zones and special events and inform approaching vehi-
cles of congestion and warning signs to reduce speeds.
3. Harris-Stephens Corner Detection and
Point Tracking
In this section, Harris-Stephens corner detection algo-
rithm used to determine interest points in the image is
discussed. Point tracking using deterministic methods for
point correspondence is used to track the vehicles. Also,
spatial and temporal characteristics are used to derive
vehicle counts. Speed of the vehicle is determined using
vector mapping and scaling of interest points at different
frames.
Harris-Stephens corner detection algorithm is based on
the auto-correlation function of a signal, where the local
auto-correlation function measures the local changes of
the signal with patches shifted by a small amount in dif-
ferent directions. The Harris-Stephen corner detection
method was improved upon Moravec’s corner detector.
The main drawback of Moravec’s corner detector is that
it is not isotropic [6]. Harris-Stephen corner detector con-
siders the differential of the corner score (auto-correla-
tion) with respective to the direction, instead of using
shifted patches.
Let us consider an 2-D image I, with an image area


,
x
y shifted by


,
x
y

. The weighted Sum of Squ-
ared Difference (SSD) or auto-correlation between the
two image patches are denoted as is given as:

,Cxy



 
2
,
,.,- ,
xy
Cxy
wxyI xxyyI xy






The shifted image I


,
x
xy y

 can be approxi-
mated by using Taylor expansion as follows:






,,,
xy
,
I
xxyyIxyxIxyyIxy 
where Ix and Iy are the partial derivatives of I with respect
to x and y respectively.
Therefore, the auto-correlation function can be ex-
pressed as an equation and as a matrix as follows:

N. CHINTALACHERUVU, V MUTHUKUMAR
308
 
2
,,.,
xy
xy
C xywxyx IxyyIxy,

 







,(
T
CxyxyA xy )

The matrix A (Harris matrix), captures the intensity
structure of the local neighborhood and is a
smooth circular Gaussian window defined as follows:

,wxy



22 2
2
,e
xy
wxy
 

The Harris matrix is expressed as:

2
2
,xy
x
y xy xy
II
I
Awxy I
II













A corner point is characterized by a large variation of
C in all direction of the vector


,
x
y.
Let λ1 and λ2 be the eigenvalues of the matrix A. By
analyzing the eigenvalues of A, the following inferences
can be made:
1) If 10

 and 20

 then the pixel

,

x
y has an
auto-correlation function that is flat and has no interest
point.
2) If 10

 and 2

has some large positive value,
then the pixel auto-correlation function is ridge shape
and interest point is an edge.
3) If1

and 2

are both large positive values, the
auto-correlation function is sharply peaked and the inter-
est point is a corner.
Since the exact computation of the eigenvalues of the
matrix is computationally expensive, computation of the
function R has been suggested by [6]. R is also refereed
as the interest point confidence value.
 
2
1212 detRAκtrace A
   

The above expression reduces the problem of deter-
mining the eigenvalues of the matrix A to evaluating the
determinant and trace of the matrix A to determine the
interest points or the corner points of the object/vehicle.

1
trace A

   

det 2
12
Aαβ λλ
λ

The interest points are marked by thresholding R and
applying non-maximal suppression. The value of has
been determined empirically, and in literature a range of
0.04 - 0.15 has been suggested.
κ
Object or vehicle tracking can be formulated as the
correspondence of the interest points across frames. In
this work, we employ a deterministic method for corre-
spondence of interest points. Deterministic methods for
interest point correspondence typically define a cost func-
tion. The cost function is a cost of associating each object
or vehicle in framesand jjk

using a set of motion
constraints. Minimization of the correspondence cost is
usually modeled as an optimization problem. However,
for vehicle tracking application the correspondence cost
is modeled as combination of proximity and common
motion constraints. In our work, the correspondence cost
employed involves matching of the object or vehicle
centroids within lanes along with spatial proximity and
common motion constraints [4]. In other words, based on
the vehicle direction the interest points move along a
common direction and the relative shift of the interest
points are uniform and can be found in certain proximity.
If


,
j
ii
C
x
y denote the set of interest points or corner
points determined by Harris-Stephen corner detection
algorithm for the frame j, then the centroid of the object
is determining as follows:

,,
ji
ii
i
xyx iy i




M
(1)
Considering the proximity and common motion con-
straint assumption the centroid approach is suitable for
determining the center of the object of the vehicle in the
bounding region or vehicle detection zone.
Our next formulation explains the approach of deter-
mining the speed of the vehicle. Let N be the number of
frames/sec captured for video processing. Let rbe the
total shift in the centroid of the vehicle from frame j to
j + k expressed as pixels. The centroid point is denoted as
d


11
,
j
M
xy and


22
,
jk
M
xy
 for the frames and j
jk

respectively. The centroid displacement in pixel is
determined as:

2
21 21v
dxxyy
2

(2)
If D (in miles) is the real-world distance from the re-
ference points on the screen to

11
,rx y


22
,rx y
d
,
then r is also evaluated by the above equation. The
speed of the vehicle that has a centroid displacement v
for the frames to
d
jjk

is determined by the following
formula:
3600 mph
vr
vdDd kN

 (3)
Figure 1, shows the reference points ( and

11
,rx y



22
,rx y), centroid displacement () and the vehicle
centroid at the framesand.
r
d
jjk
4. Vehicle Detection and Tracking
The process of vehicle detection and tracking in this
work is implemented using Harris-Stephen Corner detec-
tion algorithm to determine the corners points and the
interest points are tracked between video frames using
deterministic interest point correspondence method.
Based on the location and displacement of the interest
points, vehicle counts and vehicle speeds are determined.
The tasks employed to determine the above process is
explained as follows:
Capture of live video feed: Live video feeds from

N. CHINTALACHERUVU, V MUTHUKUMAR 309
CCTV cameras monitoring freeways and arterials are
captured for video frame processing by a USB frame
capture device. The video feeds are captured at various
locations at different time of the day. Also, the CCTV
cameras are pan-tilt-zoom cameras with varying camera
field of view. Also, the height of the camera mounted is
unknown.
Pre-processing of video frames: Using a GUI tool
developed as part of our vehicle detection system, the
user can select the region of interest on the captured vi-
deo frame. The detection and tracking algorithms are
only performed on this cropped image region to reduce
the processing time of the system. The user is required to
specify detection and speed zones using horizontal vir-
tual reference lines. The detection zones are areas where
the interest points are evaluated, vehicles detected and
vehicle counts are incremented. The speed zones are ad-
jacent to the detection zones, where the interest points
are re-evaluated and vehicles are detected. As a rule of
thumb, the detection zone length should be less than the
vehicle length as seen in the video feed and the speed
zone length should be just greater than the vehicle length
as seen in the video feed. The user specifies the virtual
vertical lane reference lines that segment the lanes on the
video frame. These vertical lines are used to determine
vehicle counts by lane. Also, the user specifies the direc-
tion of vehicle motion or traffic flow, the calibration re-
ference line and the corresponding distance in physical
distance. This reference distance is used to evaluate the
speed of the vehicle.
Smoothing: Due to low quality of image captured from
CCTV cameras (320 × 240 pixels), smoothing of the
image to eliminate noise is performed. Gaussian smoo-
thing is preferred as the noise or the nature of the object
detected could be of a Gaussian probable function. The
ROI in each frame is convoluted using a 2-D circular
Gaussian function and its discrete approximation shown
below:

22
2
2
24542
491294
11
,e 51215125
115 2π
491294
24542
xy
Gxy






 











*
(4)
Color conversion: This task converts the color im-
age/frame in the Region of Interest (RIO) from color values
to gray-scale values. The video frames captured by the
frame grabber device are in additive RGB color format.
The grayscale image is derived using the following for-
mula:
**
intensity0.2989 red0.5870 green0.1140 blue
Vehicle detection using Harris-Stephen corner in-
terest points detection: The corner points of the vehicle
in the detection zone is determined by the above men-
tioned Harris-Stephen corner point detectors [6]. The
corner points are used to detect the vehicles to determine
vehicle counts. A thresholding scheme based on pro-
ximity is used to determine the interest points belonging
to the same vehicle. If the interest points are in the thre-
shold proximity but are in different lanes, then the inter-
est points are considered disjoint. The centroid of these
corner interest points are calculated and marked in the
frame using the Equation (1). Using the centroid loca-
tions, traffic flow parameters like, total vehicle count, ve-
hicle count/lane, total volume and volume/lane can be de-
termined.
Vehicle Tracking and Speed Calculations: Using the
HS corner detector and thresholding interest points and
the centroid of these interest points are determined be-
tween detection and speed zones. The detection and speed
zones are placed adjacent to each other. Care should be
taken not to place them too far from each other and the
zones are also smaller compared to the size of the vehicle
seen in the image frame. This approach also assumes that
there exists no large velocity change of the vehicle. The
centroids of the interest points between frames are de-
termined using point correspondence method discussed
above. Since the direction of the vehicle travel or flow
has been specified using user interface, matching of cor-
respondence points (centroids) are based on proximity
and smooth motion constraints. Correspondence of cen-
troids in respective lanes is considered. This approach
will result in false detections and speed if vehicles
change lanes at the detection/speed zones. The pixel dis-
placement r) of the vehicle centroid are determined
across frames (Equation (2)) and the speed of the vehi-
cle is determined using the formula in Equation (3). Based
on the above process the system collects average speed
of the vehicles and average speed of the vehicle/lane.
d
k
5. Implementation
This section discusses the implementation of the vehicle
detection and tracking algorithm based on the Harris-
Stephen corner detector and interest point correspon-
dence discussed in Section 3. The vehicle detection and
tracking algorithm was used to determine vehicle counts,
speed and volume of vehicle in a roadway segment and
by lane. The developed system was used as a vehicle
detector for real-time advance warning ITS system. The
video frames from live video feeds from CCTV cameras
were captured using the Hauppauge Live USB frame
grabber device. The video frames were captured at (N)
29.95 frames/sec. The vehicle detection and tracking
algorithms were developed using VC++ development plat-
form using OpenCV image processing library. The video

N. CHINTALACHERUVU, V MUTHUKUMAR
310
frame captured size is 320 × 240 pixel in dimension. The points are determined (Figure 2(a)). Interest points are
further determined using non-maxima supression and thre-
sholding (Figure 2(b)). The interest points are shown by
the bounding boxes and the centroid point shown as red
dot in Figure 2(c). The bounding boxes denote the pro-
ximity region to determine if these interest points are of
the same vehicle. The centroid of the vehicle is deter-
mined by the formula in Equation (1) and a vehicle count
incremented or validated if a set number of interest points
are detected in the detection zone as shown in Figures
3(a)-(c). Vehicle speed is determined by determining the
displacement of the vehicle centroids across three video
frames
–
jk,andjjk

as shown in Figures 4(a)-(c).
The speed of the vehicle is determined as discussed in
Equation (3).
region of interest selected by the user is by default 270 ×
160 pixels.
The calibration process is supported by a user interface
that converts the real-world geometry to frame or pixel
geometry. The user interface provides the user to specify
the units of the distance reference line in feet, direction
of vehicle flow, horizontal lines to specify the detection
and speed zones, vertical lines to specify the lane demar-
cations. Figure 1 shows a snapshot of the virtual calibra-
tion lines and detection zones on the captured frame. In
order to offset the varying point of view or angle of view
of the CCTV camera, the lane demarcation lines are dou-
ble lanes for our application. The cropped frame is con-
voluted with the Gaussian function as shown in Equation
(4) to minimize the noise in the video frame. Using HS
corner detector the corner detector method the corner The developed vehicle detection system was evaluated
on a series of video feeds (8 sets) of 1 min. interval that
was recorded at many locations around the Las Vegas
valley. The video feeds vary by location, illumination
(recorded during different times of the day), road di-
mension, camera view angle and region of view. The
system was evaluated for vehicle count and speed. Vehi-
cle speeds during video recording were determined using
radar technology. Vehicle counts were manually veri-
fied. Table 1 summarizes the average count of vehicles
and the average speed of the vehicles in each lane over
all sets of evaluation video. Table 2 and Figure 6(b),
summarizes the total average speed of the vehicles for all
sets of evaluation videos.
Figure 1. Calibration lines for count and speed evaluation.
(a) (b) (c)
Figure 2. Detected Harris-Stephen corner interest points on vehicles.
(a) (b) (c)
Figure 3. Detected of Harris-Stephen corner interest points on vehicles at different lanes.

N. CHINTALACHERUVU, V MUTHUKUMAR 311
(a) (b) (c)
Figure 4. Detected of Harris-Stephen corner interest points on vehicles at frames: j − k, j and j + k.
Figure 5. Advance warning system implementation at test site.
Table 1. Comparison of speeds and counts.
SPEED COUNT ERROR
L# H A H A T H A
L1 62.6 65.5 117 124 112 5 12
L2 64 62.3 174 186 171 3 15
L3 64.4 63.7 241 219 197 44 22
L4 65.3 59.3 250 221 208 42 13
L#: Lane Number, H: HSCM, A: Autoscope, T: True manual count.
Table 2. Comparison of speed.
SPEED H A
SET 1 64.08 62.70
SET 2 64.23 64.02
SET 3 63.45 61.51
SET 4 64.55 65.37
SET 5 65.05 61.73
SET 6 64.00 63.20
SET 7 64.00 64.59
SET 8 63.48 59.40
Average 64.10 62.81
6. Vehicle Detection Application for ITS
The developed video based vehicle detection system was
employed for advanced warning of congestion and
queues at work zones and on freeways during special
events. The advance warning system consists of a series
of video monitoring stations equipped with video re-
cording devices and our video based vehicle detection
system. Vehicle queue lengths, speed and counts were
monitored before work zones or special event locations
and real-time information regarding congestions were
transmitted using Radio Frequency (RF) modules with
directional antennas to a portable variable message sign
trailer few miles downstream. Figure 5 shows the ad-
vance warning system implementation at one of our test
sites. The evaluations of the system were conducted at
various times of the day and the vehicle speeds evaluated
with and without the advance warning system in play.
The significance of the advance warning system on vehi-
cle speeds by lane is shown in Figure 6(a). Most of the
evaluations at work zones were performed during the
night and at special events during the day. The figure
shows that the advance warning system has a positive
impact on the commuters. Tere was a general reduction h

N. CHINTALACHERUVU, V MUTHUKUMAR
312
(a) (b)
Figure 6. (a) Effectiveness of advance warning system on vehicle speed; (b) Graphical comparison of the HSCM method and
autoscope for vehicle speed.
of 5 miles/hr in speeds and less traffic congestion at
work zones and special events due to the deployment of
the advance warning system. The next section discusses
the performance of the detection algorithm and the im-
pact of the advance warning system.

7. Results and Discussions
The advance warning system was developed, imple men-
ted and evaluated using many off-the-shelf components
(cameras, digital video recorders, RF communication mo-
dules, etc.), and the developed video based vehicle detec-
tion system. The initial video based detection system em-
ployed (Autoscope) requires calibration, contrast adjust-
ments, and fine-tuning of configuration parameters (ca-
mera height) for accurate results. Therefore, a video based
vehicle detection system was developed using the Harris
Stephen Corner detection method (HSCM) to eliminate the
need of complex calibration and contrasts modifications.
The performances of HSCM and Autoscope are compared
for vehicle speed and count. The performance of HSCM
is better when compared to the Autoscope with respect to
vehicle speed. The HSCM provides an average speed of
64 mph compared to 62 mph determined by the Auto-
scope. Earlier speed test using radar devices indicate that
the Autoscope determined speeds 5 mph less than the ac-
tual speed. Therefore, HSCM provides a better accuracy
for speed than Autoscope.
The performance of HSCM is better than Autoscope
for vehicle counts in Lanes 1 and 2 (lanes closer to the
camera). But for Lanes 3 and 4, the vehicle counts de-
grades significantly. This is due to the skew of the ca-
mera field of view. As the camera is installed on the light
pole in the median, there exists a considerable skew of
the captured video that results in elevated vehicle occlu-
sions. This results in some count errors in vehicle detec-
tion on respective lanes. Also due to the camera field of
view skew, some interest points of certain vehicles in
lanes 2 and 3 are detected in both lanes (lanes 2 and 3 and
lanes 3 and 4 respectively). This results in a single vehi-
cle being counted in both lanes (lane 3 and 4). The above
problem can be rectified by adjusting the camera field of
view to minimize occlusion of vehicles on adjacent lanes.
Proper placement of the virtual vertical lane reference
lines and employing camera calibration methods or trans-
forms will also reduce the count errors in lanes 3 and 4.
Lanes 1 and 2 produce better results as they are less af-
fected by the camera skew. Autoscope performed better
for this type of video, as it adopts background subtraction
method rather than interest point detection method for
vehicle detection. Future efforts will focus on improving
our current vehicle counts in lanes 3 and 4 using the above
discussed solutions. Also, shadow elimination algorithms
will be employed for vehicle detection and classification.
Finally the contribution of this work can be summa-
rized as follows: 1) a vehicle detection and tracking sys-
tem is based on interest point detection and tracking us-
ing Harris-Stephen corner detector and point correspon-
dence for developed, 2) the vehicle detection and track-
ing system is capable of determining vehicle counts and
vehicle speeds, 3) the system can determine vehicle counts
and speeds from low resolution video feeds in real-time
under various illumination conditions with very little con-
figuration and calibration requirements, and 4) the vehi-
cle detection system was used as part of an advance
warning Intelligent Transportation System (ITS).
REFERENCES
[1] Autoscope—Video Detection Solutions.
http://autoscope.com
[2] Iteris—Innovation for better mobility.
http://www.iteris.com
[3] R. Hall, “The Case for Automated Incident Detection in
Tunnels,” Iteris White Paper, 2011.
[4] A. Yilmaz, O. Javed and M. Shah, “Object Tracking: A

N. CHINTALACHERUVU, V MUTHUKUMAR 313
Survey,” ACM Computing Surveys, Vol. 38, No. 13, 2006.
[5] H. Moravec, “Visual Mapping by a Robot Rover,” Pro-
ceedings of the International Joint Conference on Artifi-
cial Intelligence, 1979, pp. 598-600.
[6] C. Harris and M. Stephens, “A Combined Corner and
Edge Detector,” The 4th Alvey Vision Conference, Man-
chester, 31 August-2 September 1988, pp. 147-151.
doi:10.1023/B:VISI.0000029664.99615.94
[7] D. Lowe, “Distinctive Image Features from Scale-Inva-
riant Keypoints,” International Journal of Computer Vi-
sion, Vol. 60, No. 2, 2004, pp. 91-110.
[8] J. Shi and C. Tomasi, “Good Features to Track,” IEEE
Conference on Computer Vision and Pattern Recognition,
Seattle, 21-23 June 1994, pp. 593-600.
[9] K. Shimizu and N. Shigehara, “Image Processing System
Used Cameras for Vehicle Surveillance,” IEEE Second
International Conference on Road Traffic Monitoring,
London, 7-9 February 1989, pp. 61-65.
[10] A. Kuehnel, “Symmetry Based Recognition of the Vehi-
cle Rears,” Pattern Recognition Letters, Vol. 12, No. 4,
1991, pp. 249-258. doi:10.1016/0167-8655(91)90039-O
[11] C. Goerick, N. Detlev and M. Werner, “Artificial Neural
Networks in Realtime Car Detection and Tracking Ap-
plications,” Pattern Recognition Letters, Vol. 17, No. 4,
1996, pp. 335-343. doi:10.1016/0167-8655(95)00129-8
[12] N. Srinivasa, “A Vision-Based Vehicle Detection and
Tracking Method for Forward Collision Warning,” IEEE
Intelligent Vehicle Symposium, Malibu, 17-21 June 2002,
pp. 626-631.
[13] C. L. Wan, K. W. Dickinson and T. D. Binnie, “A Cost-
Effective Image Sensor System for Transport Applica-
tions Utilizing a Miniature CMOS Single Chip Camera,”
Proceedings of the IFAC Transportation Systems, Tianjin,
1995, pp. 151-156.
[14] P. G. Michalopoulos and D. P. Panda, “Derivation of Ad-
vanced Traffic Parameters through Video Imaging,” IFAC
Transportation Systems, Chania, 1997.
[15] N. Hoose, “Impact: An Image Analysis Tool for Motor-
way Analysis and Surveillance,” Traffic Engineering Con-
trol Journal, Vol. 23, No. 4, 1992, pp. 140-147.
[16] N. Paragios and R. Deriche, “Geodesic Active Contours
and Level Sets for the Detection and Tracking of Moving
Objects,” IEEE Transactions on Pattern Analysis and Ma-
chine Intelligence, Vol. 22, No. 3, 2000, pp. 266-280.
doi:10.1109/34.841758
[17] B. Gloyer, H. K. Aghajan, K.-Y. Siu and T. Kailath, “Vi-
deo-Based Freeway Monitoring System Using Recur-
sive Vehicle Tracking,” Proceedings of IS & T/SPIE Sym-
posium on Electronic Image: Science & Technology—Im-
age and Video Processing, San Francisco, 5 February-5
March 1995, pp. 173-180.
[18] T. Kalinke, C. Tzomakas und W. V. Seelen, “A Texture-
Based Object Detection and an Adaptive Model-Based
Classification,” IEEE International Conference on Intel-
ligent Vehicles, Vol. 2421, No. 1, 1998, pp. 143-148.
[19] U. Handmann, T. Kalinke, C. Tzomakas, M. Werner and
W. Seelen, “An Image Processing System for Driver As-
sistance,” Vol. 18, No. 5, 2000, pp. 367-376.
[20] T. Ito, K. Yamada and K. Nishioka, “Understanding Driv-
ing Situations Using a Network Model,” IEEE Intelligent
Vehicles, Detroit, 25-26 September 1995, pp. 48-53.
[21] U. Regensburger and V. Graefe, “Visual Recognition of
Obstacles on Roads,” IEEE Intelligent Robots and Sys-
tems, Munchen, 12-16 September 1994, pp. 980-987.
[22] A. Bensrhair, M. Bertozzi, A. Broggi, P. Miche, S. Mous-
set and G. Moulminet, “A Cooperative Approach to Vi-
sion-Based Vehicle Detection,” IEEE Intelligent Trans-
portation Systems, Oakland, 25-29 August 2001, pp. 207-
212.
[23] H. H. Nagel and W. Enkelmann, “An Investigation of Smoo-
thness Constrains for the Estimation of Displacement
Vector Fields from Image Sequences,” IEEE Transac-
tions on Pattern Analysis and Machine Intelligence, pp.
565-593.
[24] B. K. P. Horn and B. G. Schunck, “Determining Optical
Flow,” Artificial Intelligence, Vol. 17, 1981, pp. 185-203.
doi:10.1016/0004-3702(81)90024-2
[25] W. Enkelmann, R. Kories, H. H. Nagel and G. Zimmer-
mann, “An Experimental Investigation of Estimation Ap-
proaches for Optical Flow Fields,” In: W. N. Martin and J.
K. Aggarwal, Eds., Motion Understanding: Robot and
Human Vision, Kluwer, Dordrecht, 1987, pp.189- 226.
[26] R. Kories, H. H. Nagel and G. Zimmermann, “Motion
Detection in Image Sequences: An Evaluation of Fea-
ture Detectors,” Proceedings of International Joint Con-
ference on Pattern Recognition, Montreal, 30 July-2 Au-
gust 1984, pp. 778-780.
[27] M. Fathy and M. Y. Siyal, “An Image Detection Tech-
nique Based on Morphological Edge Detection and Back-
ground Differencing for Realtime Traffic Analysis,”
Pattern Recognition Letters, Vol. 16, No. 12, 1995, pp. 1321-
1330. doi:10.1016/0167-8655(95)00081-X
[28] V. Kastrinaki, M. Zervakis and K. Kalaitzakis, “A Survey
of Video Processing Techniques for Traffic Applica-
tions,” Image and Vision Computing, Vol. 21, No. 4, 2003,
pp. 359-381. doi:10.1016/S0262-8856(03)00004-0
[29] P. G. Michalopoulos, “Vehicle Detection Video through
Image Processing: The Autoscope System,” IEEE Trans-
actions on Vehicular Technology, Vol. 40, No. 1, 1991, pp.
21-29. doi:10.1109/25.69968
[30] P. G. Michalopoulos, R. D. Jacobson, C. A. Anderson and
T. B. DeBruycker, “Automatic Incident Detection through
Video Image Processing,” Traffic Engineering & Con-
trol, Vol. 34, No. 2, 1993, pp. 66-75.